Training and Development Managers!

PUZZLED about where to find content for your learning program, portal or LMS?

Get 1000+ management and soft skills training resources for your staff from Mind Tools Corporate.

Find out more

Quick Tools

Monte Carlo Analysis

Bringing uncertainty and risk into forecasting

Numerous chance events.

So you've created a carefully-researched cash flow forecast for the business you want to start. And it looks like it might be a flier.

But how good is this forecast really? Will you make exactly the revenue forecasted? Exactly? And in this world of wildly fluctuating commodity prices, will your inputs cost exactly what you've forecasted?

If your revenue is just a bit down on what you expect and your costs are just a bit higher, is your promising business actually more likely to be a loss-making disaster?

And what of other forecasts – production forecasts, population forecasts, economic forecasts, even weather forecasts? How do you bring the sheer randomness of everyday life into these?

One way of doing this quantitatively is with Monte Carlo Analysis, an important decision and risk analysis technique.

Understanding the Tool:

The name "Monte Carlo Analysis" refers to the casinos at Monte Carlo in Monaco, where hundreds of chance events happen every day.

The idea behind the technique is that you feed random numbers representing the uncertain inputs (revenue and commodity prices in the example above) into your forecasting model, and then track what comes out at the other end.

This wouldn't get you anywhere if you fed in truly random numbers, or if you used only one set of inputs. However, when you generate pseudo-random inputs based on an appropriate probability distribution, and feed enough of them into your model, you generate a useful and highly informative distribution of recorded outputs.

Before PCs became common in the workplace, this was impractical. Just preparing one run of a cash flow forecast (with one set of inputs) would take a day's manual calculation and checking. However with computers, you can run hundreds or thousands of sets of inputs through your model in only a few minutes.

To do this, you'll need MS Excel Risk Analysis plug-ins like @Risk, RiskAmp or Crystal Ball, or you'll need some form of Monte Carlo Simulation front end for the modeling system you're using.